The Impact of Imperfect Maintenance Scheduling on the Physical Degradation of Painted Renderings

Abstract

Theoretical studies suggest that maintenance improves the state of conservation of buildings and components, increasing their durability and service life. However, the existing studies point to a qualitative improvement somewhere between “as-good-as-new” and “as-bad-as-old”. This study proposes a methodology to estimate the impact of imperfect maintenance actions on the degradation condition of painted renderings in external walls, based on an extensive fieldwork survey evaluating the timing and the effect of different maintenance actions on the life cycle of these claddings. Façades are analyzed at different instants (before and after the execution of maintenance services). The results reveal that the maintenance actions usually improve the condition of the painted renderings, although they are performed at inadequate moments. Cleaning actions reduce approximately 13% of the overall degradation condition of the claddings (12.2% for renderings and 13.2% for painted surfaces). The partial repair of renderings leads to a reduction of approximately 70% of the façade’s overall degradation condition. The façade only returns to the condition “as-good-as-new” when a total replacement of the rendering is performed. This study contributes to a better understanding of the phenomenon and to a more accurate projection of the real effect of the maintenance actions on the durability of façades.

1. Introduction

Buildings and their components are built to fulfil the users’ requirements for many years. Over time, the natural aging process, the exposure to environmental degradation agents and the use conditions promote the modification of the buildings’ properties, compromising their performance and the ability to accomplish adequate conservation levels. The operation and maintenance activities are responsible for 50% to 80% of the life cycle costs according to Bull and Blanchard et al. [1] and [2,3,4] and, therefore, should not be an unplanned, random or casual service, i.e., without the adoption of technical and rational criteria for the scheduling and planning of the maintenance activities. The maintenance of buildings should be considered as a programmable and technical service and as an investment in preserving the real estate assets’ value [3,5,6,7,8,9,10,11,12].

Once it is put into service, buildings and their components begin a natural process of degradation. The scarcity of resources and, in some countries, the still incipient maintenance culture, also promote the degradation of the built environment [12,13,14,15,16,17,18]. Haagenrud [19] referred to this phenomenon as “build and let decay” in the sense that there is no proper maintenance during the use phase, which results in a gradual loss of performance over time. Dann and Cantell [20] corroborated the idea with the wrong thinking of “if it is not broke, why fix it?”. Wood [21] related the little attention given to maintenance with the famous cartoon Cinderella, in which the character was initially placed in the background, but was, in fact, the protagonist in the story. Similarly, in civil construction, the attention of stakeholders is usually directed only to the design and execution phases, disregarding the relevance of maintenance for the buildings’ performance and the sustainability of the built environment.

If the maintenance policies to be adopted during the building’s life cycle were planned, with pre-established performance levels, the overall costs could be optimized, increasing the users’ satisfaction levels based on the knowledge of the building’s behavior in service conditions [16,22,23]. Silva, Brito and Gaspar [12] emphasized that the knowledge of how buildings degrade over time is the most effective way to optimize maintenance actions. According to Madureira et al. [22], the planning of maintenance actions is only possible after analyzing the performance of systems, elements and components in real conditions of use, identifying the anomalies more commonly identified and their probable causes and using degradation models to predict the components’ service life. The degradation of a building does not occur uniformly as it is composed of different systems, elements and components which degrade at their own pace as they are subjected to different demands and degradation mechanisms [12,23,24]. The buildings’ façades are among the different constructive systems and components that require regular maintenance actions throughout their life cycle. The ABNT [6], for example, suggests that façades are maintainable, although they encompass high maintenance and/or repair costs, with a replacement cost that could be higher than the initial cost.

Due to the socioeconomic situation in Brazil, the most recent buildings were built at a higher rate when compared to the constructions built a few years earlier, which in some situations, implies less rigor and control of materials and services, and disregards construction times. This situation caused a gradual decline in the quality of Brazilian constructions, promoting an increase in the incidence of anomalies from the design, installation, execution, assembly, use or maintenance, giving rise to the need for constructive diagnoses by qualified professionals (engineers or architects).

The technical diligence, named pathology, configures the systematic study of anomalies that occur during the building’s life cycle to understand the probable causes, symptoms and possible repair techniques. Generically, the term construction pathology is used in a holistic way to understand a building through the knowledge of how it is designed, constructed, used and changed, as well as the variations of materials and environmental conditions that may interfere with the building’s performance throughout its service life [25,26]. In an analogy to the medical context, the building is examined and investigated in detail, considering its age, health (degradation condition) and lifestyle (environmental exposure, use and maintenance conditions).

Therefore, ensuring an adequate performance of buildings is the responsibility of various parties, namely the designer, the constructor, the developer and the users, either during or after the expiration of the legal warranty [6,27,28,29,30]. Attia et al. [31], for example, grouped nineteen different parties into three main categories: project team, builder and customer. Furthermore, Gaspar [9], Pereira, Brito and Silvestre [32], Pires, Brito and Amaro [33] suggested that the cause-and-effect relationships of anomalies, in general, are not bi-univocal, i.e., the existence of only one defect associated with only one cause is not common. The occurrence of anomalies of different origins makes the diagnosis more difficult, because, over the years, they can hide or mask their origins, but not their appearance. Given this difficulty, the ABNT [34] considers the impossibility of framing the origin of any non-conformity during inspection services. In fact, De Paula Ramos, Petersen and Jacques [35] verified that the role of each stakeholder in maintaining an adequate performance of buildings has not been adequately addressed in Brazilian courts of justice.

Rehabilitation and maintenance actions are key factors in the sustainability of buildings [36]. The absence or inefficiency of maintenance policies favors the degradation of buildings and their components at an early stage of their service life. With proactive and preventive strategies, the physical and/or functional change in the elements can be mitigated. Furthermore, the adoption of regular maintenance strategies allows to avoid the appearance of some anomalies, minimize corrective maintenance actions and, consequently, generate less user interference [3,22].

Nevertheless, most research on service life prediction does not directly consider the impact of maintenance actions on the physical degradation of buildings. The impact of “perfect maintenance” activities is easy to analyze, but the majority of preventive maintenance actions adopted in the real world are “imperfect”, i.e., the maintenance action re-establishes the degradation condition of the element to a level between “as-good-as-new” and “as-bad-as-old” [6,37,38]. To analyze the impact of maintenance actions on the durability of buildings and their components, the actions carried out over the years must be known in detail, but the lack of reliable information is one of the main barriers encountered in this kind of research [12,39,40,41,42,43]. Usually, as discussed by Ruparathna, Hewage and Sadiq [10], this topic is evaluated indirectly through planning and cost analysis. In this sense, through an extensive literature review over the past two decades, Amos, Musa and Au-Yong [44] pointed out that there is a need to develop and focus on non-financial facilities’ management performance measures, and that there is a multidimensional taxonomy to measure performance. Thus, the impact of maintenance must be analyzed in more detail in order to understand to what extent, the actions carried out alter the natural evolution of the degradation of buildings and their components.

Previous research reveals that the operation and maintenance phases have distant and rare communication with design professionals [45,46,47]. Ishak, Chohan and Ramly [48] suggested that legal consequences are lacking for defective designs. In other words, and according to Kanniyapan et al. [49], many constructive solutions adopted, mainly in exterior and visible spaces such as façades, are often created based on subjective assumptions. Kanniyapan et al. [39], Silva, Ranasinghe and Silva [40] and Wu et al. [42] suggested that there are two main barriers to the development of maintenance projects. They are the lack of information/knowledge of maintenance criteria and the lack of anomaly/maintenance data.

2. Assessment of the Degradation Condition of Painted Rendered Façades

In this study, statistical techniques were applied to model the degradation process of painted renderings. The empirical model, proposed by Gaspar and Brito [50,51] for renderings and by Chai et al. [13,52] for painted façades, was used to obtain a numerical index that represented the severity of degradation (S_w) of the inspected façades—Equation (1). Both the types of anomalies and the weights suggested by those authors were maintained. In addition, the premise that the end of the façade’s service life occurs for a severity of degradation index of 20% was maintained. This was a conventional limit, considered as the maximum degradation condition acceptable for the stakeholders, i.e., when the renderings or painting surfaces reached a severity of degradation condition of 20%, they had to be subjected to maintenance actions or refurbishment in order to comply with the performance requirements. This theoretical limit was tested in several contexts and proved to be adequate in both Portuguese and Brazilian contexts.

$$Sw=\frac{{\displaystyle \sum } (An \times kn \times ka,n)}{A \times {{\displaystyle \sum }}^{ } (kmax)}$$

(1)

where S_w represented the severity of degradation, in percentage; A_n, the area of the façade affected by an anomaly n, in m²; k_n, the anomaly multiplier factor n, as a function of its degradation level, which varied from zero to four (Figure 1), in which zero corresponded to a façade with no visible degradation and four corresponded to a façade with generalized; k_a,n, the weighting coefficient corresponding to the relative importance of each anomaly, which varied from 0.12 to 1.53 (renders) and from 0.25 to 1.50 (painted surfaces); A, the total area of the cladding, in m²; and ∑ (kmax), the sum of the weighting factors for the highest level of degradation for each type of defect in a cladding with an area equal to A. These weighting factors intended to weigh the relevance of the different anomalies in the overall state of degradation of the façade, that is, giving greater weight to anomalies such as detachment to the detriment of anomalies such as stains. More details about the quantification of the weighting factors can be consulted in [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60].

The fieldwork was carried out from in situ inspections on the façades of vertical buildings located in the city of Porto Alegre in the state of Rio Grande do Sul, Brazil. The sample collected was composed of 18 in-use buildings, although after an initial analysis, two condominiums were considered as outliers since the anomalies observed were related to errors during the construction process. The sample was, therefore, composed of 86 painted surfaces and 79 renderings. After the managers hired different companies specialized in maintenance services, and after they started its interventions, these buildings were inspected at different times, before and after each maintenance or refurbishment action. In each moment, the severity of degradation index (S_w) of the façades was estimated. Documents of the history of construction or intervention overtime were also shared and analyzed. The aim was to adapt and extend a method to investigate the effects of different maintenance or refurbishment actions in reducing the physical degradation of façades during their life cycle, and to verify if the moment in which the interventions were carried out were adequate or according to the theoretical degradation models proposed in the literature. Therefore, a methodological proposal of the impact of maintenance actions and other interventions on the service life of the buildings was tested.

3. Maintenance and Refurbishment Services Carried Out on Painted Renderings

The most common maintenance action carried out was cleaning operations (Figure 2). Some companies considered this action as a preliminary diligence to remove degraded or loose materials, therefore, being a preparation to the subsequent maintenance actions.

In some specific cases, localized brushing was performed to remove stains and biological colonization. The cleanliness influenced the quality of adhesion of the refurbishment through the new coat of paint. The cleaning operation was usually performed by pressurized water jet, with rare adaptations.

The crack sealing of renderings usually came after the cleaning operation. Two technical options were observed in field: the application of leveling putty (Figure 3) and the use of adhesive mortar (Figure 4), and both were recommended by the Brazilian standards.

The presence of cracking in painted surfaces tended to be solved through the application of a new coating, either with products of similar quality and/or with an improvement in its characteristics, such as the use of elastomeric paints. A protective coat such as water-repellent, antifungal and biocide could also be applied in specific regions of the façades. In some cases, given the occurrence of infiltrations through cracks to the internal environment, other techniques could also be used, such as the use of a polyester veil under the renovation of the paint.

Furthermore, cleaning operations were less intrusive and worked in a uniform and continuous way, i.e., acting evenly on the entire façade. The other type of maintenance intervention, partial repair of the renderings (crack sealing or removal of parts without adhesion), was localized and could be carried out on just a region of a single part of a façade.

A more intrusive intervention (Figure 3 and Figure 4) considerably influenced the aesthetic appearance of this part of the building envelope. Therefore, after intervention, the façade was usually repainted. Furthermore, the substrate on which the paint was applied, in this case the renderings, needed to be intact and without apparent anomalies for the proper functioning of the coating system.

As for cracking, the detachment of the renderings also presented a direct effect on the painted surfaces. In this sense, Figure 5 and Figure 6 demonstrate the partial and the total repair of a single façade.

In these façades, when a detachment of the renderings occurred, the paint was also replaced regardless of its degradation condition. In this sense, the contracting parties (condominium and maintenance company) agreed that the painted rendering façade had reached the end of its life cycle. The life cycle of this rendering had ended for an overall severity of degradation (S_w) index estimated above 50%, with 42 years in service.

After those maintenance and/or refurbishment actions, the façade was repainted (Figure 7 and Figure 8). With the surface cleaned and repaired, the repainting diligences started with the application of a primer coat (Figure 7a). This procedure intended to correct defects of the substrate and/or to standardize the absorption of the surface, therefore, providing durability to the painting and economy of material. It also contributed to the adhesion of the painting scheme and promoted the bilateral chemical protection between the substrate and the paint. Figure 7b shows the covering of the first coat of paint applied over the primer.

Finally, Figure 8 reveals the result, on the right, of the application of the finishing coat. At this stage, the façade had to show a uniform appearance without defects. In the few cases in which some anomalies were observed, the maintenance company was called by the contractors for a localized repair to obtain a homogeneous appearance, without visible degradation. Therefore, in this study, after this action, the severity of degradation index returned to zero (i.e., initial moment, with no degradation), which justified the fact that the degradation curves usually applied in the literature by Gaspar and Brito [50,53], Silva, Brito and Gaspar [54,55], Silva et al. [56], Lavy and Shohet [57], Moubray [58], Shohet et al. [59], Shohet and Paciuk [60,61], Shohet, Puterman and Gilboa [30] assumed that at the time of application of the coating, the degradation must be zero (i.e., the curve must start at the origin, with the abscissa and ordinate equal to zero).

In summary, painted surfaces were generally composed of a layer of primer and two or three layers of finishing paint when a smooth finishing was desired. When a textured finishing was pursued, one or two coats of a textured material were usually applied. Equally as verified in the renderings, in other construction inspected, contracting parties accorded that the life cycle of the painted surfaces had ended at a global severity of degradation (S_w) estimated above 50%, with 21 years in service without interventions.

Based on the analysis of the major actions performed in the buildings analyzed, different methods could be used to model the effects of maintenance or refurbishment services on the physical degradation of the inspected façades.

Table 1. Distribution of the sample collected with the reasons that led to the contracting of maintenance services.

Construction	Years Since the Last Maintenance/Refurbishment	Years Since the Start of Operation	Reason for the Intervention
1	6.85	6.85	Aesthetics and cracks
2	3.19	3.19	To comply with the use, operation and maintenance manual regarding washing
3	20.75	20.75	Aesthetics, cracking and infiltrations
4	20.01	40.03	Aesthetics
5	11.89	42.91	Detachment. End of life cycle
6	11.59	42.61	Aesthetics
7	15.01	40.03	Aesthetics
8	1.10	1.10	Generalized paint cracking (outlier: construction fault)
9	7.69	7.69	Cracks in the texture and on the south side there were infiltrations
10	9.01	59.04	Aesthetics and cracks
11	7.34	37.36	Infiltrations
12	17.12	17.12	Aesthetics
13	16.62	16.62	Aesthetics, cracks and infiltrations
14	15.67	20.67	Aesthetics
15	20.01	50.03	Aesthetics, cracks and infiltrations
16	4.96	4.96	Partial heterogeneity of the original painting
17	5.04	5.04	Detachment (outlier: construction fault)
18	6.88	6.88	Aesthetics, but the condominium withdrew from the intervention due to the costs

presents the buildings analyzed, the reason that justified the maintenance action and the scheduling of the beginning of the maintenance operations. The years since the last maintenance/refurbishment and the reason for the intervention were obtained by managers or landlords. In some occasions, the condominium had technical reports suggesting how and where to intervene. The years since the start of the operation were obtained through the city hall data, which is available online.

Therefore, the evaluation of the degradation condition of painted renderings began with the comparison of façades in different degradation conditions, considering the effects of already validated variables and the reduction in its severity through different maintenance or refurbishment actions. Thus, the effects of different interventions in reducing the physical degradation of the inspected façades was highlighted. Complementarily, the adoption of a graphic method to portray the correlation between the severity of degradation index and the façades’ ages, allowed obtaining, for each case study within the sample, a minimum, an average and a maximum service life period, with and without rehabilitation actions. With this second information available, the moment when the rehabilitation was carried out pointed out if it was timely or not. Since a semi-qualitative framework was adopted, and due to the high variability of the degradation phenomena, a range period of two years was considered admissible to assume that an action was timely performed in comparison to the estimated end of the service life of the inspected element. The aim was to discuss if the rehabilitation actions carried out in the sample analyzed were performed at the best cost-effective moment. In addition, this allowed comparing the scheduling adopted in practice with the values proposed in theoretical studies and assessing the impact of maintenance and refurbishment services on the deterioration condition of the façades and on their life cycles.

3.1. Simple Linear Regression

The pattern of degradation with and without rehabilitation was compared through a regression technique. A two-dimensional graph was created with S_w before on the x axis, which represented the degradation index before any maintenance action. On the y axis was S_w-cleaning, which represented the degradation index after cleaning operations. This analysis allowed verifying the dimension of the effect of the independent variable on the dependent one. Consequently, a trend line from the origin was drawn, revealing a linear degradation pattern for painted renderings. From the analysis of the estimation of S_w-before and S_w-cleaning indicators, for each one of the case studies (79 renderings and 86 painted surfaces), the average effect of cleaning actions on the façades’ physical degradation was measured and is presented in Figure 9.

For both degradation patterns, a high determination coefficient (R² = 0.9937 and 0.9948) was obtained. Considering the information of each case study available, in both scenarios, before and after cleaning operations, a ΔS_w was estimated by subtracting the value obtained for S_w cleaning from the value obtained for S_w before. This delta represented an average weight of the effect of cleaning operations on the overall degradation of the coating. In summary, the average weight of cleaning actions on the overall degradation condition of the coatings could be estimated by the relationship between those indexes. The average effect of this action in reducing the overall severity of degradation index was 12.2% (from the division between 1 and 0.8909) and 13.2% (from the division between 1 and 0.8832) for renderings and for painted surfaces, respectively. Furthermore, the descriptive analysis of the results obtained are summarized in

Table 2. Descriptive analysis of the impact of cleaning actions on the severity of degradation index.

		Renderings	Painted Surfaces
Degradation ΔSw	Standard deviation	1.69%	1.92%
	Average	1.52%	2.83%
	Maximum observed	7.95%	8.75%
	Minimum observed	0.16%	0.00%
Case studies	-	79	86
Cases between the average and the standard deviation	-	68 (86.07%)	64 (74.42%)
Cases with a reduction superior to the average ΔSw	Framing	22 (27.84%)	29 (33.72%)
	Close to avenues with a high volume of traffic	10 (45.45%)	18 (62.06%)
	South orientation	8 (36.36%)	12 (41.37%)
	Considerable period without maintenance/refurbishment	8 (36.36%)	22 (75.86%)
	Considerable period since the entry into service	18 (81.81%)	24 (82.75%)

.

In both coatings, cleaning operations led to a reduction in the severity of degradation index superior to the average ΔS_w, namely in the following situations: (i) when the coating was located in regions near to pollutants, which promoted the presence of dirt and its deposits on the façades; consequently, for these case studies (45.45% of renderings and 62.06% of painted surfaces), the cleaning actions tended to produce a more significant decrease in the overall severity of degradation index; (ii) for coatings facing south that tended to increase the level of degradation, probably due to the action of humidity that accelerated the occurrence of biological colonization; and (iii) for coatings with more than 11 and 15 years of service since the last maintenance/refurbishment, for painted surfaces and renderings, respectively, which presented high levels of degradation and whose cleaning actions promoted the improvement of the degradation condition of the façades.

Therefore, cleaning operations had a considerable impact in reducing the physical degradation of façades that had anomalies associated with pollution and humidity. The considerations above also justified why some of the façades with similar ages presented a different ΔS_w. Façades with the same age presented different anomalies, even with similar S_w values, since some presented anomalies easily corrected with cleaning, while others presented anomalies such as cracking, whose cleaning actions had no impact on the overall degradation index of the façade. Moreover, the color of the coatings may have influenced the diagnosis of the anomalies, i.e., façades with dark colors made it difficult to quantify the real extent of the dirt, either before or after cleaning, as they maintained an appearance very similar to the naked eye.

The same procedures were adopted to measure the impact of partial repairs on the degradation of painted renderings inspected. For this purpose, a two-dimensional graph was defined with S_w cleaning on the x axis and S_w after on the y axis, which represented the severity of degradation calculated after the cleaning operations and the partial repair of the renderings through crack sealing and/or refurbishment of the detached and/or not adhered areas. As for the analysis performed for the impact of cleaning operations, the impact of partial repair of the coatings could be estimated based on the increase in the predicted service life of coatings after rehabilitation. From the S_w cleaning and S_w after calculations, for each one of the 79 renderings analyzed, the average effect of partial repair actions on the physical degradation of the cleaned renderings was measured and is presented in Figure 10.

The average effect of partial repair was 71.1% (from the division between 1 and 0.5844). Having available the information of each case study for both scenarios (after cleaning actions and after the partial repair of renderings), a ΔS_w from the subtraction of these values could be estimated. This delta represented an average weight of the effect of partial repair on the overall degradation condition of this coating (

Table 3. Descriptive analysis of the impact of partial repair on the severity of degradation of renderings.

		Renderings
Degradation ΔSw	Standard deviation	6.12%
	Average	5.24%
	Maximum observed	28.50%
	Minimum observed	0%
Case studies		79
Cases between the average and the standard deviation		73 (92.4%)
Cases with the reduction superior to average ΔSw	Framing	31 (39.24%)
	Close to avenues with a high volume of traffic	16 (51.61%)
	South orientation	11 (35.48%)
	Considerable period without maintenance/refurbishment	21 (67.74%)
	Considerable period since the entry into service	11 (35.48%)

).

After the partial repair of renderings, the painted renderings were usually repainted. The effect of full repainting of the inspected façades, as in the case of total replacement of renderings, resulted in a new life cycle period for the coating, i.e., it could be assumed that the global degradation would or should be equal to zero and the period of service life was restarted.

In general, the results showed that the longer the period since the last rehabilitation action and/or the longer the coating’s age (period since the coating started its operation phase), the greater the need for intervention and, consequently, the higher the impact of cleaning actions or partial repair on the reduction in the overall degradation condition of rendered façades. Furthermore, other variables involved in the degradation phenomena also influenced the impact of rehabilitation activities, namely, a high presence of damp or pollutants that tended to promote the presence of anomalies easily removed by cleaning operations, which influenced the impact of these maintenance actions on the overall degradation condition of painted renderings.

3.2. Graphic Method for Each Case Study

Considering the estimated severity of degradation of each case study within the sample analyzed, and knowing its real period in service since the last rehabilitation (and knowing that age was the variable with the greatest impact on the explanation of the degradation condition of painted renderings), minimum, average and maximum service life values were estimated, as well as the estimated average service life considering the standard deviation for the results of the sample (

Table 4. Summary of the estimated service life and the time when the maintenance action was performed.

-		Renderings			Painted Surfaces
		-	Average Remaining Service Life at the Time of the Intervention	Average Delay Period in Relation to the Appropriate Time for Intervention	-	Average Remaining Service Life at the Time of the Intervention	Average Delay Period in Relation to the Appropriate Time for Intervention
Intervals observed by: $SL=\frac{S\mathrm{w} \mathrm{before} \mathrm{of} \mathrm{each} \mathrm{case} \mathrm{study} }{\mathrm{age} \mathrm{of} \mathrm{each} \mathrm{case} \mathrm{study}}$	Maximum (years)	46	41	9	21	11	13
	Average (years)	18	14	4	9	4	4
	Minimum (years)	3	0	0	3	0	0
Estimated service life intervals considering the standard deviation	Maximum (years)	28	24	7	14	7	8
	Average (years)	18	14	4	9	4	4
	Minimum (years)	8	4	1	5	1	0
	Std. deviation (years)	10	10	3	4	3	4

).

Table 5. Summary of the timing of the intervention in relation to the service life of the painted renderings.

-		Renderings	Painted Surfaces
Case studies		79	86
Timely intervention		15 (18.98%)	33 (38.37%)
Untimely intervention	Before the end of service life	49 (62.02%)	23 (26.74%)
	After the end of service life	15 (18.98%)	30 (34.88%)

presents the deviations between the expected time to intervene, considering the theoretical studies, and the scheduling of maintenance actions observed in practice. Regarding the adequate scheduling for the intervention, a margin of 2 years was assumed as acceptable in relation to the end of the service life of the case study. This margin for carrying out the action was considered as appropriate, mainly because the service life prediction had some uncertainty associated due to the technical procedures necessary to carry out the action (for example, requesting budgets and granting the maintenance works to a company). The case studies inspected revealed that most rehabilitation actions did not occur at the right time (only a minority of the façades inspected were subjected to timely interventions), therefore, proving that the decision to intervene was subjective, unsustainable and not cost-effective. In the sample analyzed, the estimated service life varied between 3 and 46 years, considering the most unfavorable combination of coatings’ characteristics and the most favorable combination of variables, respectively, i.e., coatings with adequate design, using more durable materials, more protected from environmental degradation agents and subjected to interventions presented higher estimated service lives. Even though, in the sample analyzed, approximately 62% of the renderings were subjected to rehabilitation before the end of their service life, with an average remaining period of performance at the time of the intervention of 16 years (considering the most favorable combination of coatings’ characteristics). On the other hand, approximately 27% of the painted surfaces were subjected to rehabilitation before the end of their service life, with an average remaining period of performance at the time of the intervention of 5 years.

4. Discussion

The results obtained through the data collected in this study revealed that:

• Different services led to distinct decreases in the physical degradation of the façades;
• Cleaning operations reduced, on average, 13% (12.2% for renderings and 13.2% for painted surfaces) of the overall severity of degradation;
• Ferreira et al. [62] corroborate these results, suggesting that the adoption of minor interventions or cleaning operations can increase operating costs, but, on the other hand, when carried out with an adequate frequency, can allow the performance of components to be maintained at high standards during most or part of their service life, thus, promoting its durability;
• Partial repair of renderings reduced, on average, 71.1% of the façade’s overall severity of degradation;
• As suggested by Aikivuori [63] and by Silva and Brito [11], among other authors, rehabilitations were predominantly carried out at inappropriate times;
• According to Table 1, Table 4 and Table 5, and other findings, there was a lack of technical basis in decision making. This probably stemmed from the lack of: (i) interaction between the design, the execution and the maintenance phases and (ii) the lack of periodic inspections and/or technically inefficient guidance/reporting;
• According to Table 5, among the 79 renderings inspected:

  ○

  In 49 cases or in 62% of the sample, rehabilitation was carried out before the end of their service life;

  ○

  In 15 cases or in 19% of the sample, rehabilitation was carried out after the end of their service life, with an average delay of 5 years;

  ○

  In only 15 cases (19% of the sample), interventions were carried out at the appropriate time in relation to the end of the element’s service life.

• According to Table 5, of the 86 painted surfaces inspected:

  ○

  In 23 cases or in 27% of the sample, rehabilitation was carried out before the end of their service life, with an average remaining period at the time of the intervention of 5 years (for the most favorable combination of coatings’ characteristics);

  ○

  In 30 cases or in 35% of the sample, rehabilitation was carried out after the end of their service life, with an average delay of 6 years;

  ○

  In 33 cases, or in 38% of the sample, interventions were carried out at the appropriate time in relation to the end of the element’s service life.

• Ferreira et al. [8] corroborate these results by suggesting that maintenance actions are carried out before the appropriate time, which leads to unnecessary costs or postpones the intervention to moments of urgency, when users’ safety may already be compromised. Cardinal et al. [64] also suggested that the inspections tended to result in reactive maintenance, and in this sense, the inspections should be performed more frequently, according to a planned maintenance strategy, to optimize the moments for interventions in a cost-effective way;
• The record of interventions carried out during the life cycle of the sample analyzed did not meet the technical guidelines of the ABNT [5,28,65]. In this sense, Kanniyapan et al. [39], Silva, Brito and Gaspar [12], Silva, Ranasinghe and Silva [40], Souza [41], Wu et al. [42] and Zanoni [43] cautioned that academic researchers and industry professionals face difficulties in obtaining data on anomalies resulting from a lack of maintenance.

5. Conclusions

Studies on the impact of rehabilitation on the degradation condition of buildings and components are at an infancy stage, with limited real practical testing and data. This study intended to fill this gap in the existing studies, based on theoretical assumptions. In general, the results obtained in this study confirmed that the effects of intervention were relevant to reduce the physical degradation of the inspected façades. Cleaning actions reduced approximately 13% of the overall degradation condition of the claddings (12.2% for renderings and 13.2% for painted surfaces). The partial repair of renderings led to a reduction of approximately 70% of the façade’s overall degradation condition. Furthermore, rehabilitation was always beneficial and allowed instantaneously reducing the façades’ degradation, as well as increasing the durability of the coatings. From this reduction in the severity of the degradation over time, verified through its patterns, it could be inferred that the service life of the coatings could be increased and, consequently, allow a more sustainable and economical management throughout the life cycle of the painted renderings.

The study showed that the anomalies rarely occurred in a generalized way, since different variables demonstrated different performances over time. In fact, renderings and painted surfaces presented specific points of degradation, and when these degradation mechanisms were not contained or mitigated, the pattern curves showed that the degradation process tended to accelerate, promoting a continuous deterioration of contiguous regions.

Painted renderings were the most common type of external coating and their degradation had visible consequences for the citizens’ perception of the cities. In this sense, the rehabilitation performed should not have compromised the aesthetic appearance of the façades. The results of this study revealed that the maintenance and the refurbishment actions on renderings and paintings tended to be integrated and carried out simultaneously to ensure that the façade presented a homogeneous appearance after the intervention.

Another relevant conclusion of this study was the verification that, in addition to validating the theoretical concepts of rehabilitation through the study of buildings in service, the actions occurred, in their vast majority, at inappropriate times, either before or after the end of their service life. For renderings, in only 19% of the sample, the interventions were carried out at the appropriate time in relation to the end of the element’s service life. However, the results revealed that the decision to intervene in the painted surfaces was more precise or accurate (38% of the sample) than the decisions made for the renderings. The results pointed out that the rehabilitation performed in painted renderings was usually trigged by the aesthetic appearance of the painted surface, neglecting the overall degradation or its functional effects. This solution seemed partially reasonable, since, among the two coatings studied, paintings were less durable with a lower estimated service life.

In the sample analyzed, when the painting required a maintenance action, the entire façade was repaired, perhaps due to an “opportunity cost” of contracting and negotiating with the maintenance companies. Therefore, the decision to intervene was based on programmatic criteria rather than on technical criteria related with the physical degradation of the coatings. The actions carried out that were inspected were often only partial, not repairing the rendering adequately, allowing it to continue to degrade and having to be repaired, refurbished or replaced at the end of its service life or its life cycle. In this sense, no concerted measures were taken to opportunistically repair the two coating systems, optimizing their maintenance costs.

The adoption of degradation models to evaluate the impact of maintenance and other strategies over time provided very useful data to aid stakeholders during the decision to intervene, in order to optimize funds and resources. The results of this study can aid some design decisions, for example, knowing that the durability of paintings with a predominant southern orientation in the southern hemisphere is lower, therefore, suggests the necessity to specify more durable paints than those applied on other façades or even another type of coating (such as ceramic claddings) that can improve the average performance of the building envelope, standardizing the overall performance of the building and reducing the number of interventions and the rehabilitation costs. This is relevant because, in most of the situations, the ordinary expenses of a condominium (concierge, sanitation, water consumption, electricity, etc.) did not include planned rehabilitation costs. In this sense, knowledge to adopt more rational and cost-effective strategies is crucial to optimize the resources for routine, preventive and technically desirable interventions for the proper performance of buildings and their components.

This study, thus, suggests the following strategies in future, to promote a more rational use of resources during the buildings’ life cycle, reducing the unnecessary costs while increasing their performance and durability, namely: (i) new studies that directly analyze the effects of rehabilitation on buildings in service; (ii) new studies that directly analyze if rehabilitation on buildings in service are timely carried out; complementary, (iii) adopting design and maintenance plans, considering that the buildings and their systems, elements and components present different performances, service life and life cycle ranges (use degradation models to establish more technically informed strategies); (iv) including information about the maintenance and operation phase during the design and the execution process; and (v) creating new ways to record information about the maintenance actions over time, either using BIM (building information modelling) or using other forms of digitization of information about the buildings’ operational phase.